AI Heartbeats

Computers have learned to predict survival in patients with a heart condition more accurately than is possible today. In the future, this may help doctors to identify patients at greatest risk of death earlier and treat them more intensively. The artificial intelligence software was developed using historic data from over 250 patients with pulmonary hypertension, which can lead to heart failure. The software analyses moving images of a patient’s heart captured during an MRI scan. It then uses image processing to build a 'virtual 3D heart', replicating the way that over 30,000 points contract during each beat. By linking its models with the patient data, it learned which attributes of a heart, its shape and structure, put an individual at a given risk of heart failure. Declan O’Regan, who led the research at the MRC LMS, says, “The computer is up to 80% accurate at predicting survival at one year.”

Read more about this research here.

Written by Deborah Oakley

Deborah Oakley

Writer

Deborah is researching healthy cognitive ageing at the University of Cambridge. She previously worked as a science communicator in London, working closely with the BPoD team.

• Video by Declan O’Regan, MR Facility, MRC LMS, London, UK
• MRC London Institute for Medical Sciences, London, UK
• Video copyright held by the authors
• Research published in Radiology, January 2017